publishable final activity report 8

IP 017498DYNAMITE

Dynamic Decisions in Maintenance

Publishable FinalActivity Report 8

M1-M42September 1st, 2005 – February 28th, 2009

Version 1.0

Edited byKenneth Holmberg & Irina GranforsVTTApril 15th , 2009

Integrated Project funded by the European Community under the 6th Framework Programme,Priority 2 IST and 3 NMP joint call

Contract Start Date: 01.09.2005 Duration: 42 monthsProject coordinator: Kenneth Holmberg, VTT Technical Research Centre of Finland

DYNAMITE EU 6FP IP 017498

Final Public Project Activity Report 8, M1-M42, V 1.0 / 15.3.2009 2

DYNAMITEPublishable Final Activity Report 8

Deliverable dataDeliverable no& name

Activity Report 8

MainContributors

Kenneth Holmberg & Erkki Jantunen (VTT), David Baglee (SUN), Aitor Arnaiz(TEK) and Julien Mascolo (CRF)

OtherContributors

Basim Al-Najjar (VXU) and Benoit Iung (UHP)

ShortDescription

Publishable Final Activity Report of DYNAMITE, period 1.9.2005 –28.2.2009

Disseminationlevel

Draft restricted to DYNAMITE partners, will become public afterapproval

Date 15.4.2009Status DraftInternalreview by:

All partners

Externalreview by:Internallyaccepted by:

All partners

Date ofacceptance:

14.4.2009

Document historyVersion Date Author /

ReviewerDescription

V 0.1 9.3.2009 VTT – KH Model delivered to contributors 9.3.2009V 0.2 9.4.2009 All partners Sent out to all partners for check

9.4.2009V 1.0 15.4.2009 VTT – All partners Final, submitted to EC 15.4.2009



DynaWeb e-maintenance solution for dynamicdecisions in maintenance

1 IntroductionMaintenance is the field of technology that consists of technical skills, techniques,

methods and theories which all aim at technical and organisational solutions for largeassets like factories, power plants, transportation vehicles and constructions, as wellas for smaller assets such as household machines, hobby devices and consumerproducts, to function properly, in a cost effective way, with low energy consumption,without polluting the environment and in a safe, controlled and predictable way. Thecosts and risks related to improper maintenance are huge. Poorly functioningproduction machines and unreliable products are negative for a company’s business.Effective maintenance is crucial for the competitiveness and the future of a company.

Figure 1.1 The DYNAMITE vision for future IT-based maintenance.

The optimal solution would be to know continuously the condition of the asset andits components and take repair and service actions only when really needed. Todaythere is an improved understanding of the phenomena initiating and triggeringdisturbances and failures and new effective sensor as well as information andcommunication techniques that makes it possible to analyse and deal accurately withlarge amount of data and information on global level (fig. 1.1). The concept of e-maintenance that takes advantage of this development is defined as “The networkthat integrates and synchronizes the various maintenance and reliability applicationsto gather and deliver asset information where it is needed" (Baldwin, 2001).

We report here a flavor of advanced techniques and methods that form the basisfor a novel DynaWeb e-maintenance approach and solutions including advancedmicro sensors, smart tags (RFID, radio frequency identification), on-line oil sensors,PDA maintenance applications, ontology based diagnostic and prognostic methods,wireless communication, semantic web service for distributed intelligence, dynamiccost effectiveness based decision making tools and a holistic e-maintenanceconcept. Experiences both from laboratory testing and use in industrial environmentsis presented. The reported cases are demonstrations on global level, with millingmachine, machine tool, foundry hydraulics, maritime lubrication system andautomatic stamping machine.



2 DYNAMITE project objectives and contractorsThe DYNAMITE project aimed at delivering a blend of leading-edge communicationsand sensor technology, combined with state-of-the-art diagnostic and prognostictechniques, which will advance the capabilities of European industry in maintenance.The objective was to deliver a prototype system with a clear exploitation route to takethe technology into the European market.The monitoring of machines and processes for predictive maintenance and control iscrucial for sustainable and competitive industry in Europe. Distributed, autonomousmonitoring is fundamental to the penetration of e-maintenance to the cutting edge ofhigh capital and highly productive plant. DYNAMITE aimed at creating aninfrastructure for mobile monitoring technology and creating new devices which willmake major advances in capability for decision systems incorporating sensors andalgorithms. The key features were planned to include wireless telemetry, intelligentlocal history in smart tags, and on-line instrumentation.Objectives1. A flexible infrastructure to host technologies for global e-maintenance

2. A hardware and software demonstrator of smart tag technology incorporatingidentity, health history and communications

3. A mobile device for access to and reporting from the e-maintenanceinfrastructure

4. A low cost wireless sensor demonstrator – a self-powered, multi-sensor, smartdevice with wireless communications

5. An online oil sensor incorporating particle counting, distribution and compositionin one instrument

6. Tools and methods for cost-effective applications of the maintenancetechnologies for continuous enhancement of company’s profitability andcompetitiveness.

ContractorsThe DYNAMITE contractors are listed below.

No. Code Name Country1 VTT VTT Industrial Systems Finland2 TEK Fundación Tekniker Spain3 SUN University of Sunderland UK4 MAN University of Manchester UK5 UHP Université Henri Poincaré France6 VXU Växjö University Sweden7 ZEN Zenon S.A. Robotics & Informatics Greece8 CRF FIAT Research Center Italy9 VOL Volvo Technology AB Sweden10 GOR Goratu Maquinas Herramienta S.A. Spain11 WYS Wyselec Oy Finland12 MAR Martechnic GmbH Germany13 ESS Engineering Statistical Solutions Ltd UK14 DIA Diagnostic Solutions Ltd UK15 PRI Prisma Electronics Greece16 IBK IB Krates OÜ Estonia17 HYD Hydrox Pipeline Oy, 1.9.2005-28.2.2007 Finland



3 E-maintenance scenario analysisThe design of a flexible structure for DYNAMITE (Dynamic decisions in maintenance)is supported on the assumption of the existence numerous companies that canbenefit from a subset of the technologies addressed in the project, providingcustomised plug and play to the desired upgrades with respect to each company’sexisting maintenance activities. It is also understood that there is not a single‘upgrade’ solution that fits for all concerning the maintenance needs.

Given this, one of the first activities performed in the project, apart of a conventionalstudy of requirements for ICT and sensor technologies, has been the study of the usecases involved in the project, with the aim to identify clear separate scenarios fordemonstration of new technologies, which should also facilitate the implementation ofcost-effective maintenance solutions. Scenarios are extracted out of initial use cases,as likely representative examples of a wider group of companies having similarobjectives in the maintenance process, sharing similar technology status, or sharinga need with respect to maintenance technologies needed. As a result, different havebeen separated, including large companies with de-centralised production, OEMsuppliers, small companies with few dozens of machining systems and third partyconsultants.

Table 1, which was compiled from end user analysis during the first stages ofDYNAMITE, clearly shows that our initial assumption is true, and it is not possible tofind a single system for a global upgrade of existing maintenance systems. Theexisting strategies, legacy systems and other issues differ very much, as well asperceived technical problems and economical motivations. However, if we take thistable as reference, it allows generalizing scenarios that can go beyond a particularuse case, and thus provide an entry point to the technologies for companies thatshare similarities with one of the scenarios (Roles, operational contexts, applications,components, preferred upgrades etc).

Table 1: Summary of end-user scenarios in DYNAMITE project

Providers Plant operators OEM manufacturers Consulting Trasport (OEM +consulting.)

ContextSingle location

(Manufacturing plant).Multiple machines

Technical AssistanceServices. Guarantees

Multiple locations

Specialised services(e.g. lube analysis) for

multiple locations

Specificmachinery on

movementApplication

(Components)Milling, drilling and high speed machine tools

(Hydraulic systems, gearbox, spindle)Motors (Marine,

Automotive)

Current strategy New PM ( 10- 20 %CBM) BDM PM -

Currenteconomic

motivations

Overall economicimpact to the company

for differentmaintenance strategies

not always known

Uneven workloadEnforce/surveil

remote maintenanceprocedures on

guaranteedmachinery

Decrease downtime,repair and

maintenance costs

Plan new cost –effective e-

Maintenance fornew equipment

Currenttechnical

problem(s)

Evaluation of machinecondition depending on

expert knowledge(subjective)

Improve diagnosisCommunicationsensors to OEM(bypass CNC)

Lack of experienceddiagnosis and

decisions over existingparameters

Lack of properknowledge

Interestingtechnologies

Include advancedsensorsWireless

communication Smart PDAs Include cost-effectiveness

Upgrade to CBM Useremote monitoring

e-Maintenance.Wireless gatewaysCost effectiveness

Use e-Maintenance toremotely assess

expert andcommunicate to

operators. Trainingsystems.

Initiate predictivemaintenance

Likely CBM/PMparameters(sensors)

Temperature, Voltage, Current,Oil level, Oil quality, vibration, pressure, Wear debris



4 Intelligent sensors

The development of intelligent sensor is fundamentals to support different diagnostic,prognostic and maintenance activities. In Dynamite project, three categories ofintelligent sensors are focused including smart tags, micro sensors and lube sensors.

Three categories of intelligent sensors contain different purpose. The investigation ofsmart tags is concentrating on utilising existing radio-frequency identification (RFID)technology to improve the assets maintenance and management. Different to thesmart tags, the role of investigating micro sensors is to develop a more powerful andself-powered wireless micro-electro-mechanical systems (MEMS) sensor foradvanced condition monitoring. And finally, the investigation of lube sensors istargeting on various sensor techniques for analyzing and detecting differentlubrication features.

For smart tags, both passive RFID and active RFID technology are considered. Thepassive RFID is suggested to be used for the replacement of barcode system and itis recommended specific for asset identification and inventory purpose includingmachines, spare parts and tools. Moreover, passive RFID and PDA can be usedtogether as a perfect maintenance tools. It can effectively reduce the improper assetidentification in order to prevent a series of inappropriate maintenance activities.

Alternatively, the active RFID is perfect for real-time location tracking (RTLS) ofmobile assets. It can be applied for security and mobile assets tracking purpose todetect any unauthorized people getting into a protected area, and also search andreserve any shared mobile resources like vehicles.

For micro sensors, a powerful wireless micro-electro-mechanical systems (MEMS)sensor for advanced condition monitoring has been developed. It combines andintegrates three sensors including vibration, temperature and pressure together anda Zigbee wireless communication module to a single MEMS sensor. By using this,multiple sensor values can be collected and transmitted directly to the wirelesssensor network for processing at the same time.

Also, in order to support the idea of wireless senor technology continuously workingfor a long time, a self-powered electronic power management module has beendeveloped to bridge the wireless sensor and the harvesting devices including solar-based devices and vibration-based devices for recharging batteries.

In the Dynamite project, four types of lube sensors are focused in the lubricationsystem: fibre optic laser and SLED absorption and scatter sensors for solidcontaminants, particle senor, oxidation sensor and water sensor. Various sensortechniques described here are important in analyzing and detecting differentlubrication features

Firstly, a low cost fibre optic laser and super light emitting diode (SLED) absorptionand scatter sensors for solid contaminants sensor uses optical fibres as data andenergy channels to give well a cleanliness index for solid particle content in themeasured lubrication.



Secondly, a particle sensor is developed to measure particle content of lubricating oil.Through using a CCD camera together with an illumination system, particles evensmaller than 1 micron size can also be detected.

Thirdly, an oxidation sensor is used to measure oxidation level of lubricating oil.Through measuring the transmittance of the light in the visible range (380-780 nm) ofthe light spectra, the degradation status of the lubricating oil can be calculated by thecorrelation of absorption of light and the oxidation level of lubricating oil.

Fourthly, a water sensor can measure water content of lubricating oil. Similar tooxidation sensor, the transmittance of light in the NIR range (about 1400 nm) of thelight spectra can be measured and check with the correlation the absorption of lightand the water content of lubricating oil.

The following figure illustrates a complete information flow of a lubrication system.Various lube sensors are connected to the target system for data collection. Then aPC or a PDA can be used as a basic data collector to extract features and criticalinformation and send the results to the Mimosa database for storage and furtherprocessing.

Figure 1. The complete information flow from lube sensors to Mimosa database

In order to support the manipulate of sensor data in the Mimosa database, acomplete set of web-services are designed and developed to support assetinformation querying, monitoring, diagnostics and prognostics. Those web-basedservices and sensor techniques are detailed as different DYNAWeb components.

Finally, it is necessary to notice that all works described in this section includingsmart tags, wireless MEMS sensors and lube sensor systems are successfully testedin laboratory and demonstration. And they are fully support the connection to theMimosa database for storage and retrieve of information. Based on that, engineerscan call different web services provided to read the up-to-date data anywhere byusing their PC, PDA and mobile phone via accessing WiFi and mobile internet.



5 Information and communication infrastructure

Actual industrial maintenance activities are mainly driven by traditional strategieswhere events are normally time-based. Tough it has been pointed out that moreadvanced, condition-based strategies can provide clear savings in manymaintenance activities, their application is normally prevented by different causes,such as the need to manage, both physically and logically, an increasing volume ofdata and information. At DYNAMITE, one of the main developments has been relatedto the development of a flexible architecture concept to provide flexible data andinformation management.

On the one hand, a platform of web services to provide intelligent processingcapabilities has been designed. This platform is logically structured according toOSA-CBM decision layers, from Condition Monitoring to Decision Support, but also tothe existing operators (Sensors, PDA, CMMS etc). The result is a three levelframework (machine, plant, company) that provides a flexible configuration thatallows the system to be used ‘on-demand’ and to grow according to the needs (newsensors and functionality), together with a flexible communications infrastructure,where a generic wireless ‘gateway’ device is being developed, in order tocomplement existing communications options (wired or wireless) between sensorsand company decision areas when other communication options are not available(such as SCADA, PDAs, etc). The different components developed, together withadditional issues related to a global approach were firstly envisaged as a globalsystem that was nicknamed DynaWeb. The structure is highlighted in the followingfigure.

Serviceproviders

Equipmentmanufacturers

Enterprise servicesplatform (ERP)

Enterprise assetmanagement (EAM)

BlackBox

Zigbee

Ethernet (WIFI)

Web services overinternetBusiness softwareMIMOSA DB compliant

CMOpS (LAN)Stores reliability andcondition monitoring information(Failures, parameters,…)

CMMS/ERPMaintenance ExpertStores work order andmaintenance strategies information

Ethernet

Wired(RS-232,RS485, CANLAN, ..)

RF

SensorsTemporal

Smart tagsPermanent machine identification

Management/Offices

Production& MaintenanceOperation

Data processor(RS232, LAN)

PDA (Zigbee, WLAN)Partial storage. ObservationsMaintenance execution

Machine

DataCollector

CommunicationBridges

Zigbee

(RS232, RS485,Zigbee, WLAN)

Ethernet

BluetoothRF

Figure 2: DynaWeb platform

On the other hand, this platform has been populated with different ICT componentsto facilitate an this web distributed e-maintenance solution



A mobile handheld device has been developed, and includes wireless access tosmart tags and sensors and centralized databases within the e-maintenanceinfrastructure, including application software for performing analysis of monitoringdata and early stage diagnose of faulty conditions. Specific developed modules thatcan be pointed out include the interfaces for managing the information andcommunicating with operators, the infrastructure and agents for interoperation withremote web services, and the inclusion of specialized models to retrieve informationfrom smart tags and optimise the maintenance scheduling

A dual system for wireless communication across different operators has beendeveloped. This system is composed of a physical gateway for communicationbetween machine and plant, plus a data collector to facilitate communicationbetween wired systems and wireless gateways. The protocols used are Zigbee andWifi. This physical system is complemented by a translator developed to assist theinterpretation of SQL Queries into MIMOSA database structure.

A complete set of web services have been designed and developed, covering thecomplete OSA-CBM information process layered structure. These services canoperate from any type of location (sensors, PDAs, PCs) and provide an standardizedmeans to access to information located in any machine, without the need to changeexisting legacy systems, with a minimum need of adaptation.

Last, it can also be stated that MIMOSA arquitecture was decided to be a central partof our development process. This has allowed different positive outcomes:

- Interoperability between all developed components have been secured notjust along the across the entire project

- Many components developed can now be used in two different ways: Straightto MIMOSA databases, or via XML message passing to non-MMOSAdatabases, using also MIMOSA architecture in the protocols. In this way, theweb services can also be standardized, no matter where is the informationstored.

- The use of MIMOSA standard, even tough not much extended yet at thebeginning of the project, seems to become aware in many different areas,which now turns into a positive selling argument, as the compliance of anysystem with MIMOSA is allowing interoperability among an increasing numberof maintenance software systems.

All the results achieved have been detailed in a list of DYNAWeb components.Concerning ICT communications the available components resulting fromDYNAMITE are shown below

For direct operation of the PDA

2. Active smart tag, asset tracking Zenon9. Mobile maintenance PDA user interface VTT10. Smart tag PDA support Sunderland University11. Active smart tag PDA support Zenon12. Hand held PDA vibration data collector Diagnostics Solutions Ltd.13. PDA scheduling support Zenon14. Smart PDA maintenance user interface Prisma technologies



For communications across different operators (sensors, PDA, PC), MIMOSAdatabase and web services

15. Communication SW module Prisma technologies16. Mimosa translator University Henri Poincaré17. Collector (=Gateway) Prisma technologies18. Wireless communication system for e-maintenance Prisma technologies

Concerning web services development

19. Condition monitoring web service TEK20. Diagnosis web service TEK21. Prognosis web service University Henri Poincaré22. DynaWeb e-maintenance platform (TESSNet) Fundación Tekniker23. Scheduling web service Zenon27. MEMS SW support module Diagnostics Solutions Ltd28. Vibration measurement system Wyselec

Concerning structures for MIMOSA

25. DynaWeb platform IBK26. Mimosa database IBK

6 Cost effectiveness based decision supportIn general, decisions of when and why to stop a producing machine and whether it iscost-effective or not are crucial for production profitability especially in companies ofintensive capital investments, e.g. process industry, shipping and engineeringmanufacturing, where stoppage time is very expensive. It is vital to have a systemproviding reliable data required to achieve cost-effective and dynamic maintenancedecisions for maintaining and improving company profitability and competitiveness.

A novel Maintenance Decision Support System (MDSS) was developed that offersthree different strategies for cost-effectiveness that can be applied integrated orseparately, Fig.6.1. MDSS can help companies to reduce economic losses throughmapping the situation of production and maintenance processes and enhancemaintenance performance. It allows following up maintenance performancemeasures more frequently, thereby be able to react quicker on disturbances andavoid unnecessary costs. It will also be easier to trace the causes behind deviations.

MDSS consists of three toolsets, where every toolset consists of one to three toolswith different functions, see the Table 1. MDSS provides services that existingsystems cannot do. It helps to identify the most beneficial areas for futureinvestments in maintenance. MDSS applicability and usefulness have been testedsuccessfully by personnel from FIAT/CRF (Italy) and Goratu (Spain) and it has beeninstalled at Fiat/CRF, Italy for test since 16th of Jan. 2009. The final conclusion of thetest and demonstration of MDSS is that MDSS is user friendly and can be usedsuccessfully for analysis of data and achievement of maintenance dynamic and cost-effective decisions.

Table1. Functions of the toolset and tools included by MDSSToolsets Tools

Features & Function

MDSS:Toolset 1Toolset 2Toolset 3

Easy to use, effective and low cost



Toolset 1to enhance theaccuracy ofmaintenancedecisions

A. PreVib(Prediction ofVibration level)

To predict the vibration level of a component/equipment in the next plannedmaintenance action or measuring moment. It is important to avoid sudden anddramatic changes in the vibration level and catastrophic failures. Prediction isdone using specially developed Mechanistic model that re-assess itsparameters and constants after each vibration measurement using non-linearregression

B. ProFail(Probability ofFailure)

To assess the probability of failure of a component (using machine past data)at need or when its vibration level is significantly high.

C. ResLife(ResidualLifetime)

To assess the residual life of a component. To avoid failures and deliverydelays, ResLife can be used to control whether it is possible for the productionprocess to proceed according to the production schedule or not. Theprobability of failure and residual life are assessed using a modified form ofTotal Time on Test (TTT-plots)

Toolset 2simulate andselect the mostcost-effectivemaintenancesolution

A. AltSim(AlternativeSimulation)

To simulate technically applicable alternative solutions suggested for aparticular problem, and to select the most cost-effective maintenance solutionusing an intelligent motor. This tool is important to improve the cost-effectiveness of maintenance investments and planned actions. We use herethe same sets of formulas used for MainSave for converting technical impact ofmaintenance on the company business to economic measures. The selectionis done using well-defined criteria, such as the proportion of maintenancesavings to the invested capital, total maintenance profit.

Toolset 3to identify &prioritise problemareas

A. MMME (Man-Machine-Maintenance-Economy)

To identify and prioritise problem areas and to assess the losses in theproduction time. It is beneficial to plan maintenance actions according to aprioritising list justifying maintenance actions. A special model for analysingproduction process with respect to production time losses and the causesbehind is used in this tool.

to map, follow up,analysis andassess the cost-effectiveness ofmaintenance

B. MainSave(MaintenanceSavings)

To monitor, map, analyse, follow up and assess maintenance cost-effectiveness, i.e. maintenance contribution in company profit (maintenancesavings & profit). It is a reliable tool for securing cost-effective maintenanceactions. A set of formulas for converting technical impact of maintenance toeconomic impact on the company business are used for Main Save.

Fig.6.1. The MDSS Maintenace Decision Support System of strategies for cost-effectiveness.

16-17 June 2008 Växjö WP12-Meeting

MaintenanceDecision SupportSystem (MDSS)

MaintenanceDecision SupportSystem (MDSS)

Toolset 1Accurate

MaintenanceDecisions

Toolset 1Accurate

MaintenanceDecisions

Prediction ofVibration Level

(PreVib)

Prediction ofVibration Level

(PreVib)

Assessment ofProbability of

Failure (ProFail)and

Residual Lifetime(ResLife)

Assessment ofProbability of

Failure (ProFail)and

Residual Lifetime(ResLife)

Toolset 2Analysis Tools

Toolset 2Analysis Tools

AlternativeSimulations

(AltSim)

AlternativeSimulations

(AltSim)

SystemSystem

ToolsTools

ToolsetsToolsetsToolset 3

Cost-effectivenessToolset 3

Cost-effectiveness

Man-Machine-Maintenance-

Economy (MMME)

Man-Machine-Maintenance-

Economy (MMME)

MaintenanceSavings

(MainSave)

MaintenanceSavings

(MainSave)



7 DynaWeb integrated solution

In order to solve software related technical issues and to integrate the work inDynamite project a Software Team was nominated. The software team discussedissues related to the programming tools and techniques used in the project and triedto unify the programming work and support the communication between variousmodules developed by various partners in Dynamite.

The Software Team members created a document/deliverable that describes theprogramming techniques used in programming the software modules of Dynamiteproject “D7.6 A Short Tutorial on How to Build Web Services, Agents and PDASoftware”. Even though the title suggests that the document is not big in size it in theend became a 130 pages long document containing the most important aspects andguidelines for programming Dynamite modules. During this process of definition ofthe basic rules for programming it was realised that the challenge on how to organisethe communication between various modules of Dynamite had not been solved in theDescription of Work (DoW) i.e. the project plan. After long discussions in a separatemeeting the decision was made to rely on a common database when integrating thevarious software modules. The decision was not easy to make especially becausemany of the partners already had some maintenance related software together withvarious database formats. However, it was realised that here lay the key to successof Dynamite i.e. it could only succeed if all the modules could communicate togetherand therefore a common ground was needed. In fact the whole idea in e-Maintenance is to be able to pass information to where ever it is needed so thecorrect maintenance action can be taken at the right time using effectivemethodology.

When the decision had been made to rely on a common database for data exchangebetween the Dynamite software modules it was a logical decision that the databaseshould be Mimosa. Why to choose Mimosa? Mimosa organisation gives the followingdefinition “MIMOSA is a not-for-profit trade association dedicated to developing andencouraging the adoption of open information standards for Operations andMaintenance in manufacturing, fleet, and facility environments. MIMOSA's openstandards enable collaborative asset lifecycle management in both commercial andmilitary applications.” Clearly this is the optimal strategy for an international projectaiming for collaborative work and integration of results.

As such Mimosa is relatively big containing hundreds of tables. Mimosa definitioncovers issues related to measurements, condition monitoring, diagnosis, prognosisand management of maintenance work orders etc. Rather soon after the adaption ofMimosa it became clear that it was not an easy step for such partners that were notused to working with relational databases. Even though Mimosa is well documentedand it is easy to download and install to run e.g. under SQL Server it is not an easystep to start using a database in a logical way. In fact quite a lot of effort was spent indiscussing how Mimosa should be used in order it to be an effective tool.

The common Mimosa database was installed in the project server at IB Krateslocated in Tallinn. IB Krates also built a clever user interface to help the use ofMimosa and especially the manual input of data into Mimosa. After using Mimosa fordata exchange all partners by the end of the project agreed that the decision to go forMimosa was the right decision to make and that in fact no other solution is seen tohave had a similar effect in supporting the integration within Dynamite. Figure 7.1



shows the communication within DynaWeb in simplified flowchart format. As can beseen in Figure 7.1 Mimosa database is the central point where most of the data goesand where it can be read from. As can be seen there is also communication betweenvarious modules on module to module basis following the same common dataformat.

Figure 7.1 Simplified flowchart of the communication within DynaWeb throughMimosa database.

8 DynaWeb demonstrations

DynaWeb demonstrations were carried out in industrial environment on globallevel, with milling machine, machine tool, foundry hydraulics and a maritimelubrication system.The functionality of the DynaWeb and its components were tested and demonstratedin the following way:

demonstrations were performed in 4 different test sites at Fiat, Volvo, Goratuand Martechnic,

technical and economical evaluations of these demonstrators were carriedout and

recommendations for further implementation, development andindustrialisation were done.

The following results were achieved on the four demonstration sites:

1) Fiat tested and demonstrated the integration between 25 DynaWeb hardware,software components and services. The demonstration was done in an industrialmachining centre similar to what is used in car production. Detailed results areavailable in the technical reports. As a summary of the demonstration:

The overall results are extremely positive, with technical and economicalfeasibility proven.



The level of quality of components and adequacy to requirements was high,with people extremely dedicated to enhancing their components and testingthem onto the demonstrator.

As expected, integration was not straightforward and required a major effortfrom all involved partners.

Unfortunately some components were not delivered on-time and thus notintegrated.

2) Volvo tested the oil sensor system from Tekniker designed to measure the level ofoxidation of the lubricant by spectroscopy of visible light. The demonstration wasdone in a hydraulic system in real industrial environment, production line in thefoundry. In conclusion:

The oxidation sensor hardware and software worked fine in the foundryinstallation.

The environment in the foundry at Volvo was extremely dirty which was agood test for the sensor but made it impossible to have the computer at thesame location. The sensor required a continuous low speed oil flow withoutair bubbles and at a low oil pressure. The sensor signal was jumping up anddown depending on e.g. irregular oil flow, air bubbles etc. and made theinterpretation more difficult and not straight forward.

Volvo IT policy made it almost impossible to demonstrate communication withthe Mimosa database at external server but a one-way web servicecommunication to store data in the Mimosa database was created byTekniker and included in the software and tested finally.

3) Goratu tested several DynaWeb components and their communication to theMimosa Database.

The VTT Particle Scatter lube sensor for hydraulic system, the TEK watercontent lube sensor for cooling system and the Wyselec vibrationmeasurement system for spindle vibration were implemented at a Goratumachine, and the data collected where sent to the Mimosa Database locatedat IBK server. All the data collected provided great information for Goratu,who did not have any kind of information related to these issues. Apart fromthis, the web services provided an important tool for machine reliability. Webservices allow Goratu to implement the online diagnosis and conditionmonitoring, which was impossible until now.

The hand held vibration unit and the PDA maintenance user interface weretested too. These allow inserting new assets on the database and doingmeasurements using a PDA, which due to its size is very comfortable for theuser.

This demonstration gave new feedback to Goratu, who will use thisinformation for machine improvements and new utilities for the customers.The innovation is very important, but some improvements are needed for afull implementation in an industrial environment, higher flows and pressure forsensors and better filters for vibration system.

4) Martechnic, the demonstration consisted of a simulated application of a stern tubebearing/tail end shaft assembly from an 8000TEU container ship. For logistical andsecurity reasons this demonstration could not take place on board the ship. Speciallydesigned test rig conditions on board a ship were replicated and the cycled lube oilwas progressively contaminated with water and particulate matter.



The demonstration which ran for nine days evaluated four sensors (one fromTekniker was not enabled for DynaWeb communication). The other three (twofrom Martechnic and one from VTT) performed satisfactorily communicatingtheir results via two separate routes to the MIMOSA database.

The demonstration was deemed a success and the economic scenariosurrounding this application clearly demonstrated considerable benefits fromapplying the DYNAMITE concepts.

9 Standardisation issuesAt the beginning of the DYNAMITE project some standards already existed that couldinfluenced on the investigations and developments to be done. The main standardswere EN13306 Maintenance Terminology, ISO/DIS18436-2 & ISO13374 (conditionmonitoring purpose); IEEE 802.11x and IEEE 802.15 (communication purpose);EN457:1992- ISO7731:1986, EN418:1992, EN60204-1:1997/IEC60204-1 (Safety ofmachinery); IEC6224; MIMOSA OSA/CBM, IEEE 1232 (interoperability purpose).Among these standards, MIMOSA (Machinery Information Management OpenSystems Alliance) was used to support the integration issues in DYNAMITE as thekey pillar. The interest for the DYNAMITE components to be MIMOSA compliant is toensure that information can be freely exchanged with all other MIMOSA compliantapplications without the necessity of any customized interpreter or wrapper.The results issued from this compliance and detailed in most of the deliverables ofDYNAMITE, represent potential material for current standardisation initiatives in twomain aspects:- MIMOSA-based initiatives to develop standard in Maintenance area to support

integration issues. Indeed MIMOSA is not really today a standard approved byall but initiatives advocated by International Organisation are referenced toMIMOSA for normalising “interoperability profiles” between all maintenanceprocesses. Some modifications on MIMOSA issued from DYNAMITE, couldimpact, for example, the proposal ISO TC 18435-1 – Industrial automationsystems and integration – Diagnostics, capability assessment, and maintenanceapplications integration – Part 1 to 5.

- Wireless-based initiatives to focus on wireless monitoring and alerting needs forthe industries. Such initiatives such as ISA 100 Wireless Networking Committeewhich is progressing toward its first standard ISA100.11a, are intended toprovide reliable and secure operation for non-critical monitoring, alerting,supervisory control, open loop control and "soft" closed loop control applications.Taking into account results of tests done on PRISMA Gateway and Web-Services, DYNAMITE could impact the first thinking on ISA100.11a.

Thus it can be expected that a future challenge after DYNAMITE could be to developsuggestions to launch work at several national levels within the national standardorganisations (e.g. AFNOR in France) to push DYNAMITE results (Interoperability-Integration aspects; Wireless aspect) as a contribution to these new standards forimproving the initial version (CD or DIS format) to lead to last one (IS format).



10 Conclusions1) A pioneering e-maintenance solution named DynaWeb was developed. It is basedon scenario analysis of future industrial needs and trends for plant operators, OEMmanufacturers, transportation and consulting companies.2) DynaWeb is a flexible web distributed ICT structure capable of multilevel conditionmonitoring and maintenance data treatment with common MINOSA structure,internet web services, training services and decision support based on technical andeconomical considerations.3) DynaWeb consist of 28 integrated hardware and software components. Theyinclude smart MEMS sensors with energy harvesting, on-line lubrication sensors,smart tags for identification and location of components, maintenance actionssupporting handheld mobile computers (PDA), wireless communication and astrategic and economical decision support system.4) A condition monitoring data and statistically based decision support system MDSSwas developed. It includes toolsets for accurate maintenance decisions, maintenanceanalysis and cost effectiveness.5) DynaWeb components and the integrated structure was successfully tested anddemonstrated on global level, with milling machine, machine tool, foundry hydraulics,maritime lubrication system and automatic stamping machine industrial installations.



NEW web site http://dynamite.vtt.fi/

http://dynamite.vtt.fi/

publishable final activity report 8

Documents